The Hidden Cash Crisis in “Profitable” Online Fashion Stores

As a founder of Datahut, I've spent the last 14 years working with a lot of fashion stores, both small and large. We;ve had a front row seat as we saw real data powering decisions at successful fashion retailers. 

One of the problems I've seen with struggling stores is that they’re profitable on paper but their cash is tied to their inventory.

Am I talking about you?

Lets find out

Open your P&L and you might see enviable numbers:

• Gross margins comfortably above 55 %

• A conversion rate nudging 3 %

• Double-digit month-on-month growth

  
  

Yet your current account feels anaemic. Why? Because as much as 20 %–30 % of the value in every garment evaporates into storage, insurance, capital interest, shrinkage and markdown risk—the classic inventory-carrying cost range for fashion retail.

Layer on the industry’s average sell-through rate of just 40 %–80 % and you realise you’re effectively paying rent on your own cash. Every stagnant SKU also hurts perception: stale collections, broken size runs and perpetual clearance banners scream “bargain bin,” not “premium brand.”

Bottom line: inventory is not an asset until it moves. Until then it is a cash-eating liability. The antidote is radical visibility—turning messy data into simple, actionable numbers you can trust.

What is happening in the market 

Sell-through rate (STR) – the percentage of inventory sold within a period versus received – is a critical gauge of how well fashion e-commerce brands balance supply with demand. Industry benchmarks consider ~70–80% sell-through as the “ideal” range for healthy performance. In practice, most retailers aim to sell about 70% of a season’s stock at full price before markdowns.

Recently, however, achieving that has become tougher: initial sell-through by the start of clearance often hovers closer to ~60% in fashion, as consumers’ purchasing shifts and seasons blur. Top-quartile performers (e.g. successful fast-fashion and lean DTC brands) consistently hit 80%+ sell-through in-season, meaning the vast majority of their styles sell without discounting.

Lagging brands or marketplace sellers with poor alignment might see sell-through rates down in the 40–50% range, indicating a lot of slow-moving stock. (Industry experts note that a sell-through below ~40% is problematic, leading to heavy markdowns and write-offs.

By the end of the full cycle (including clearance sales), retailers generally target around 90–95% total sell-through so that only a minimal fraction of inventory remains unsold. For instance, luxury and premium brands – which traditionally had very high full-price sell-through – have also felt pressure; many now resort to outlet channels or online flash sales to clear stock and reach close to 90% sell-through by season’s end.

Notable case: ASOS’s recent “Test & React” initiative (a rapid small-batch program) achieved ~60% sell-through within just 7 days of launch for new products, with those items turning 3× faster than its average stock. 

This underscores how data-driven assortment and nimble restocking can dramatically lift sell-through performance, approaching the levels of an Amazon marketplace seller (many marketplace sellers consider ~80% STR as excellent, and may discontinue or liquidate items that perform far below that.

Working Capital & Inventory Carrying Costs

Efficient inventory management is crucial for freeing up cash in fashion e-commerce. Inventory ties up a substantial share of working capital – in fact, in 2022 the inventory held by global fashion companies averaged about 20.7% of annual revenue.

This ratio has risen from roughly 18–19% pre-2020 to over one-fifth post-pandemic, reflecting larger stock buffers and excess goods in recent years. In dollar terms, this means a fashion e-tailer with $1 billion revenue might have over $200 million sitting in inventory on its balance sheet. Such capital could otherwise be invested in new products, marketing, or technology. The scale of the issue is evident in aggregate: the fashion industry over-produced an estimated 2.5–5 billion garments in 2023, resulting in $70–$140 billion worth of unsold stock globally– a massive amount of cash effectively locked in inventory. 

Even top brands were not immune: luxury leaders LVMH and Kering together held about €5 billion ($5.4B) in excess fashion inventory in 2024, and Nike saw the portion of its products needing discounting jump to 44% in 2024 (from just 19% in 2022) due to inventory overhang. These examples illustrate how excess stock directly impacts working capital and profitability.

Inventory carrying cost

Inventory carrying cost is another key metric: it encompasses storage, insurance, depreciation, and the cost of capital for held inventory. Retail benchmarks show that holding inventory typically costs about 20–30% of the inventory’s value per year. In other words, a company with $100 million of inventory may incur $20–30 million in annual carrying costs (warehouse space, financing, obsolescence risk, etc.). This is why improving turnover and sell-through has outsized financial benefits – every week reduction in average stock levels saves money and releases cash. 

High-performing e-commerce fashion players leverage techniques like demand forecasting AI, just-in-time replenishment, and drop-shipping to minimize on-hand stock (thus reducing carrying cost). For instance, Shein’s ~40-day inventory model not only boosts sell-through but also sharply cuts storage time and markdown expense Similarly, many marketplace-based sellers (who often use fulfillment services like FBA) are incentivized to keep stock lean, since overstocking can raise holding costs by 25%+ and even incur extra fees

The most recent data and reports from firms like McKinsey, BCG, and BoF indicate that improving inventory metrics remains a critical priority for fashion e-commerce. Those that succeed in boosting turnover and sell-through (through better demand analytics, agile supply chains, and tighter buying) will not only free up cash and cut costs but also outperform peers in responsiveness and profitability

A Five-Step Rescue Plan You Can Run in Google Sheets

Step 1 – Export the Basics

1. Pull 12–24 months of data from Shopify, WooCommerce or your ERP:

   • Orders (with dates & quantities)

   • Stock receipts (PO dates & units)

   • Returns (date, SKU, qty)

2. Standardize your CSV:

   • Ensure consistent SKU formatting (no stray spaces or dashes)

   • Normalize date formats (YYYY-MM-DD)

   • Filter out cancelled or test orders

3. Set up your sheet tabs:

   • Raw Data

   • Cleaned Data (with formulas applied)

   • Metrics Dashboard

     
     

Pro Tip: Use Google Sheets’ “Import” functions (e.g. IMPORTDATA) or a simple n8n/Make scenario to auto-refresh this CSV each month.

Grab 12–24 months of order, stock and return data from Shopify, WooCommerce or your ERP. Save as CSV. Minimum columns are given below

SKU | Style | Colour | Size | Qty Received | Qty Sold | Date Received | Current Stock | Returns

Step 2 – Add Three Killer Metrics

• Automate your formulas so they update when you paste new data.

• Color-code high/low performers with conditional formatting.

• Add dynamic date filters to compare rolling 4-week vs. 12-month trends.

Step 3 – Find Heroes & Laggards

• Sort by ST % (desc) and WOH (asc) in your cleaned Data tab.

• Define your groups:

  • Heroes: Top 20 % (ST % ≥ 80 % & WOH < 8)

  • Watch-List: Middle 60 % (all others)

  • Laggards: Bottom 20 % (ST % ≤ 40 % & WOH > 12)

• Create pivot tables to break out heroes/laggards by:

  • Style

  • Colour

  • Size

  • Seasonality tag

• Visualize with a simple bar chart—heroes vs. laggards—so you can instantly see where cash is trapped.

  
  

Insight: Look for patterns—are certain colours or sizes consistently lagging? That’s often a sizing-curve or styling issue, not just “bad demand.”

 Step 4 – Rebalance the Next PO

1. Heroes first:

   • Increase reorder qty by 10–30% above last cycle’s sell rate

   • Consider testing a small new colourway or size on 1–2 % of total qty

2. Laggards last:

   • Pause fresh POs until you’ve liquidated ≥ 50 % of existing stock

   • Bundle or cross-sell via targeted promotions (e.g., “Buy a hoodie, get a laggard tee 50 % off”)

3. Size-curve alignment:

   • Mirror your actual sales distribution vs. vendor’s suggestion

   • Use your pivot “Size” tab to calculate exact ratios

4. Lead-time buffer:

   • Factor in supplier lead time variability—if your average lead time is 21 days ± 5 days, order 10 % extra heroes to cover delays.

     
     

Step 5 – Rinse & Repeat Monthly (Analyse → Rebalance → Test → Analyse again.)

1. Set a calendar reminder on Day 1 of each month:

   • Export fresh data

   • Re-run your metrics tab

2. Track your progress:

   • Maintain a “Month-Over-Month” tab showing changes in overall ST %, average WOH, cash released

3. Continuous testing:

   • If a new tactic (e.g., bundling) frees up > 5 % extra cash, roll it out to similar SKUs

4. Scale with minimal effort:

   • Once your Sheets workflow is bullet-proof, automate the data pull (via Datahut feeds or an integration tool) so you spend 30 minutes reviewing, not copying & pasting.

Bonus: Share your dashboard with key stakeholders (e.g., merchandising, finance) so everyone sees the impact of freed-up cash—and stays aligned on the plan.

How Web Scraping Supercharges Each Step

Here’s a more detailed look at how web scraping supercharges each step—packed with extra boosters and the tangible edges you gain:

Extra Tips:

• Cross-Market Signals: Scrape international sites to see which styles are trending globally before they hit your region.

• Supplier Health Check: Monitor vendor websites or marketplaces (e.g., Alibaba, IndiaMART) for lead-time changes or stock shortages—critical when POs are 60+ days out.

• Pricing Psychology: Track competitor price thresholds (e.g. the “.99” effect) and adjust your pricing slice to maximize perceived value.

With these data-driven boosters, your simple Sheets workflow becomes a full-blown market radar—delivering competitive edge at every turn.

Putting ChatGPT on Autopilot (Zero Data Team Needed)

Even a solo founder can mimic a BI team by pasting a CSV slice (2 000 rows tops) into ChatGPT with Advanced Data 

Analysis:

Prompt:

“Load the file and:

1. Compute sell-through %, WOH, and seasonality tag by SKU | size | colour.

2. Flag SKUs with WOH > 12 or ST % < 40.

3. Cluster SKUs into heroes (top 20 %), middle (60 %), laggards (20 %).

4. Recommend reorder quantities to keep total stock value < $ 1Million.”

ChatGPT will:

• Generate pivot tables in-memory.

• Visualise a heat-map of seasonality spikes.

• Output a cash-release forecast—e.g., “Cutting laggard buys by 30 % frees ₹18 lakh over 90 days.”

• Produce Python code you can copy into Google Colab for repeat runs.

  
  

Supercharge with Datahut: Feed ChatGPT a second CSV scraped by Datahut—competitor pricing and stock snapshots. Ask:

“Overlay my hero SKUs against competitor stock-out frequency.

Which items should I push in ads this month for maximum margin?”

Within minutes you have campaign ideas with built-in pricing leverage.

Turning Laggards Into Revenue—Without Discounting Your Brand

Deep discounts erode perception. Instead, treat laggards with:

Hero-Anchored Bundles: Pair a best-selling jogger (ASP ₹2 199) with a slow-moving tee (COGS ₹200, ASP ₹999). Frame it:

“Free tee worth ₹799 when you grab our bestselling joggers.”

AOV stays high, margins intact, tee exits quietly.

Gift-with-Purchase (GWP): Offer stale scarves as GWP on orders ≥₹3 000. Perceived value rises; shelf clears.

Tiered Loyalty Rewards: Let VIPs redeem points for laggard SKUs—moving stock while buying goodwill.

Flash Bundles via WhatsApp: Send a private drop to VIPs: “24-hour lookbook bundle—30 % off.” Urgency without public markdowns.

Metrics That Matter Going Forward

Build a lightweight dashboard in Google Sheets or pipe Datahut Feeds into Looker Studio for efficient tracking.

Author 

I’m Tony Paul ,founder of Datahut . I’ve been working in the web scraping industry for over 14 years. Working with top fashion brands and their data teams - Me and my team at Datahut learned a lot of invaluable lessons on how to use data to free up cash tied to the inventory. If you're someone facing this trouble and want to solve these problems. Get in touch with us using the chat widget on the right hand side.